Alternative therapies for drug refractory B cell malignancies are urgently needed. We used advances last cycle to develop a new class of recombinant bispecific targeted toxin with expanded ability to recognize a wider range of B cell cancers. With funding from this grant, we discovered that bispecific targeted toxins could be made by fusing two repeating scFv subunits creating a bispecific ligand directed toxin and they had superior activity over monospecific targeted toxins. We brought the drug to clinical trial for leukemia. Findings are encouraging, but preliminary data indicates that toxin immunogenicity will be a problem as anticipated. Multiple dosing is required to obtain sustained remissions and this generates anti-toxin antibodies that impact drug efficacy. To this end, we have found an important new and successful way to deimmunize our toxin that may have implications for deimmunizing other molecules. In aim 1, we propose to study this deimmunization. Funding from this proposal has also led to the development of a very different targeted toxin that has the ability to target cancer stem cells. Current theory dictates that cancer cells differentiate as do normal cells. The cancer cell population consists of a small cohort of less differentiated, self renewing, tumor initiating stem cells and a larger cohort of more differentiated tumor cells. More differentiated cells are more susceptible to chemotherapy and the minor fraction of less differentiated cancer stem cells are more drug resistant, thus contributing to drug refractory relapse in cancer. We have created a new targeted toxin that targets cancer stem cells and clearly has therapeutic efficacy in systemic mouse models of cancer. The agent has the advantage of working against a variety of different cancers. In aim 2, we propose to study this new strategy of targeting tumor stem cells. Aim 1 relates to Aim 2 since the deimmunization strategies are used to construct anti-stem cell targeted toxins as well.